bel8

6 ms·

31.
by bel8 29d ago
fair but that's also a matter of time, in my opinion. 1) software doesn't commit suicide like some pilots did. 2) software has superhuman reflexes, processing power and are not affected by emotion. 3) software also will eliminate
32.
by bel8 29d ago
> ...but for serious work, it will never be a sure shot job handled entirely by AI. Maybe not entirely handed by AI but when AI starts doing it better than most and in a fraction of the time, for a fraction of the cost, why do you think
33.
by bel8 1mo ago
The decision to not contribute when you have infinite resources is a hostile action on itself.
34.
by bel8 1mo ago
"Developers developers developers!"
35.
by bel8 1mo ago
Regardless, saying it would be too much effort for Apple is laughable at best. Even more so when Apple used to release such docs in the past when it was much less financially capable. They just don't care.
36.
by bel8 1mo ago
2legit2quit obviously, given the context.
37.
by bel8 1mo ago
The fact that they would rather try to win an internet argument by showing indifference to what others have to say than learn something might hint at traits that keep them unemployed. Harsh but maybe this helps changing their mindset.
38.
by bel8 1mo ago
It usually boils down to people trying to convince themselves that keeping their macs hot and with very little ram to spare only to get sub 50 tokens per second on a subpar lobotomized (quantized) model is worth it. And I'm not even co
39.
by bel8 1mo ago
> We were speculating that maybe it's because a lot of benches are best-of-n, and should probably be worst-of-n. Thank you. I just changed my opinion on this thanks to you. I agree now, since we tend to execute LLM tasks once instea
40.
by bel8 1mo ago
Props to your parent commenter for including context size. Because 131k context window is prohibitively small for my coding workloads so I know a 512GB Mac won't cut it.
41.
by bel8 1mo ago
true but if you're actually running k8s and similar workloads, chances are it might eat memory that LLM requires. you'll also notice these articles rarely specify their context window in tokens, because it is small, usually 30k to
42.
by bel8 1mo ago
or switch to codex
43.
by bel8 1mo ago
it's really not, I see my colleague using https://www.fontlab.com for professional font authoring. I'll ask them about glyphs, got me curious.
44.
by bel8 1mo ago
just to clarify, models already ran locally without MLX years before it existed, on non-Apple environments. MLX was just Apple's bridge to what already ran in other hardware.
45.
by bel8 1mo ago
I have been using it since it was available in opencode Go plan. It replaced all other models for me. It's much less chatty than DeepSeek V4. And it feels noticeably smarter. I also use Claude at work and GLM 5.3 feels like Opus 4.8 (w
46.
by bel8 1mo ago
isn't GLM 5.3 flash better than GLM 5.2 overall?
47.
by bel8 1mo ago
You're not sending any "data" though, much less sharing anything. I just asked my colleague and they had no clue how to do it in their iPhone. I guess a win for Apple.
48.
by bel8 1mo ago
How is "share" intuitive here? I want to pin the app, not share. That seems purposefully misplaced. As usual with malicious compliance. In Android: Menu > Install and create shortcut. Much cleaner, intentional, and visible with
49.
by bel8 1mo ago
If you're on opencode's go $10/mo plan and want to use GLM-5.3-flash right now on pi, you can add this to models.json until pi updates to support it: { "providers": { "opencode-go": {
50.
by bel8 1mo ago
Source? GLM is great for coding and Gemini is barely useful in coding, to be generous. I highly suspect that the Gemini Google uses internally is very different from what they offer in Antigravity.
51.
by bel8 1mo ago
It's about friction. Examples: Docker on Linux is miles better and efficient. Window management and Desktop environment on Linux is whatever I want it to be. Open source gives me freedom: I (LLMs) can investigate/fix/change a
52.
by bel8 1mo ago
I tried to love my work issued m4 pro for many months but couldn't stand macOS so asked for a linux compatible machine.
53.
by bel8 1mo ago
As long as you don't force everyone else to do the same, whatever works for you.
54.
by bel8 1mo ago
The real downside for me is not having Linux support. It would take Apple one or two engineers to make Linux life much easier on macs. But Linux is outside their walled garden so it's ignored.
55.
by bel8 1mo ago
What doesn't sound intellectually stimulating is grouping up all AI proponents in a single crowd. Also, the anti-AI author's app leverages a critical dependency that does use AI. So their stance pass no sniff test.
56.
by bel8 1mo ago
This feels like saying you'll never use self-driving cars. Sure, LLM effectiveness might seem debatable to some (not for me), but these things improve frequently. So saying you'll never use it, is a red flag to me and I question w
57.
by bel8 2mo ago
That's another goalpost. That's because DRM doesn't go well with Open source.
58.
by bel8 2mo ago
isn't Qwen more expensive? Much more verbose than flash on high, and and slower?
59.
by bel8 2mo ago
I don't doubt you. I just think it's good practice to run more than n=1 before conclusions are made. We're all engineers here.
60.
by bel8 2mo ago
Yes, it adds vision to the already capable text-only LLM according to DS: > This experimental multimodal model matches DeepSeek-V4-Flash on text capabilities—including agents, reasoning, and world knowledge. https://api-docs.d

More ›